Jul 18, 2025, 10:00 AM ETInterview

Tech Executives: AI Has Changed SaaS Forever (Don't Fall Behind)

Salesforce changed its fundamental pricing structure three times within a single 12-month period, breaking the historical norm of major companies adjusting pricing models only once every five years.
Large enterprises are currently moving at "historical speeds," rendering traditional nine-to-18-month case study and rollout cycles obsolete for startups and legacy firms.
The market is transitioning from three distinct monetization eras: the on-prem/perpetual license era, the cloud/seat-subscription era, and the current AI/"value" era.
The AI era shifts the value proposition from "who has access" to "what work the software performs," decoupling revenue growth from headcount scaling.
Usage-based billing is no longer exclusive to infrastructure; it is now a core strategic component of the business plan for SaaS companies.

Real-time Necessity: Unlike monthly billing, usage billing requires real-time detection of spending because API consumption can reach unbounded amounts (e.g., $1M in OpenAI) in hours, necessitating systems that prevent "rogue process" breakage.
Data Complexity: Usage billing functions as a "data infra problem married to billing complexity," requiring joins across 50+ data sources rather than simple user counts.
Dynamic Rules: While pricing models appear static, enterprise contracts often involve highly dynamic, negotiated line-item discounts that are currently handled manually in many public companies.
Data Integrity: Financial accuracy must be 100%, not 99%, as any error is considered fraud; this requires preserving raw data for future re-pricing without relying on approximations.
Technical Debt: Engineering talent often avoids billing infrastructure due to its "boring" nature and accumulation of technical debt, leading to fragile, custom-built systems that break at scale.

Incentive Realignment: Usage models theoretically align sales incentives with customer value (comp paid only upon actual usage), whereas seat models incentivize headcount growth regardless of product utility.
CEO Commitment: Successful adoption requires a CEO-level "dictator" to force compromises between Sales, Product, Engineering, and Finance, as these groups have conflicting objective functions.
Sales & GTM Restructuring: The role of Account Executives must shift from closing deals to ensuring product adoption, potentially requiring a split between pre-sales (qualification) and post-sales (adoption) functions.
Customer Success Evolution: The role shifts from expansion-focused to retention-focused, with CS teams compensated on gross churn and customer health rather than upsell metrics.
Finance as Data Org: Finance teams must operate at a weekly or daily clock speed rather than quarterly, acting as strategic data providers to ensure accurate, real-time revenue recognition and usage reporting.
Compensation Friction: Sales teams often resist retraining and new comp plans; without a centralized authority to "ramrod" these changes, the transition stalls.

Hybrid Models: The dominant near-term model is hybrid (seat fees + variable usage costs), serving as a bridge to pure consumption by anchoring revenue while allowing upside capture.
Pure Usage Segmentation: Pure usage models are predicted to dominate in B2B enterprise and AI agent layers (where value is work performed), while B2C will likely remain subscription-based to reduce cognitive load for consumers.
Market Saturation Strategy: Private companies are using fixed-margin, cost-plus pricing as a weapon to saturate the market, gain ubiquity, and fund future model development rather than maximizing immediate revenue.
Agility Over Optimization: There is no single "right" pricing model; the only rewarded trait in the current environment is the agility to experiment with pricing and packaging.
Performance Guarantees: Companies are using aggressive guarantees (e.g., Intercom offering $1M in credit if resolution rates don't improve) to validate value and prove the efficacy of usage-based pricing.

Revenue Control: In usage models, engineers have direct control over revenue; a 50% efficiency improvement in compute reduces revenue by 50%, requiring controlled rollout of optimizations (e.g., metering efficiency gains over quarters).
Core Value Metrics: Product teams must obsess over the "core volume metric" (the driver of value) as it is now literally their revenue metric, shifting product management toward explicit revenue generation.
CEO Oversight: Executives must monitor core usage metrics daily, analyzing the impact of new workloads on spend curves rather than waiting for monthly or quarterly reviews.
Historical Parallels: The current AI wave mirrors the early internet in terms of growth velocity and market consolidation, but differs in that monetization strategies (usage-based) are already understood and scalable.